Independent guide to useful Bots · Across platforms
Bot Cabinet
← Crew Kits
New crew · manual setup

Market Lookout

Know which competing products changed and whether those changes matter to your business.

Set up the crew and pass the results between Bots yourself. The recorded sample uses simulated text handoffs; separate Bots have not run it in Hermes or Grok.

Download the plan

When to use this crew

Use this crew to track competitor changes, verify them and check how products appear in searches. For a simple price comparison, Researcher may be enough.

The jobs

  1. Competitor Brief

    Compare dated public product, price, and hiring information.

  2. Researcher

    Check important claims against their original sources.

  3. Search Presence

    Check a fixed set of search and AI-answer observations.

What you will need

Named competitors, approved public pages or snapshots, prior dated records, exact questions, region, and what decision the report should help.

Set up and run

  1. Select only the roles this job needs. Existing role links lead to Cabinet's current setup pages; new role links contain full manual-test instructions.
  2. Put each selected role in its own conversation or Bot. Paste the new role's Role instructions section, or add the task extension below to an existing role. These files are not native import archives or add-to-app links.
  3. Start with the fictional first task below. Supply the same approved source records to each role that needs them. A copied handoff alone is not evidence for a factual claim.
  4. Pass reviewed results in the order below. If your app lacks Bot-to-Bot handoff, copy the small handoff yourself. Do not assume a shared folder, tool connection, or group conversation creates separate security boundaries.
  5. Record each actual answer and failed attempt. Review before connecting real accounts or adding a schedule. Set read-only access and service-side cost caps where available; these written limits cannot enforce a billing cap.

Pass the work in this order

  1. Competitor Brief returns changes with dates and source records.
  2. Researcher checks the strongest claims and challenges unsupported inferences.
  3. Search Presence returns separate, dated observations against the approved questions.
  4. Researcher combines the checked changes and Search Presence observations into one report, preserving the original source records and unknowns. The person decides whether to act on its proposed response.

Use this handoff

  • Task and current version:
  • Approved source records:
  • Result and its status: draft, checked, or approved:
  • Unresolved questions:
  • Next owner and action:
  • External action awaiting specific approval:

If a role cannot finish, it returns the missing input or failed attempt. The next role must not manufacture a success record.

First task

Fictional records: a rival's same per-seat monthly plan moves from $29 on Oct 1 to $39 on Oct 5. Its new job advert is Enterprise Sales Lead. One supplied AI answer incorrectly describes your paid plan as free. Prepare a change report and a correction request without doing new searches.

Check the result

  • Reports the $10 change without mixing annual and monthly prices.
  • Keeps the hiring advert separate from a confirmed roadmap.
  • Identifies the wrong price claim and preserves the supplied-observation label.
  • Makes no private inquiry or automatic website change.
Read the sample answer and checks

Sample written in Codex on October 6, 2026. The same assistant wrote and checked the answer. This is not an independent review or a Hermes or Grok app test.

Original answer

Text simulation only: named roles below represent staged text review in this exercise. No searches, separate Bot runs, or contacts were performed.

Change report, prepared October 6, 2026

  • Pricing: the supplied October 1 record gives $29 per seat monthly; the October 5 record gives $39 per seat monthly for the same plan. The increase is $10 per seat per month. Keep both observations as the comparison record. The original observation year, currency jurisdiction, region, plan features, and original page links were not supplied. Source: original fictional pricing records.
  • Hiring: the supplied record describes a new public advert titled 'Enterprise Sales Lead.' That is a hiring signal that may indicate investment in enterprise sales, not proof of an enterprise product launch, revenue, or financial health. No advert observation date, posting date, location, or link was supplied. Source: original fictional advert record.
  • Search presence: one supplied AI answer describes your paid plan as free, which the supplied task identifies as incorrect. This is one pricing-description discrepancy, not evidence of overall visibility or ranking. The original question, service, answer date, region, settings, full answer, and cited links are missing. Your plan is described as paid in the original task; its actual price, billing period, business name, and authoritative pricing-sheet/page excerpt were not supplied. Source: original fictional AI-observation and paid-plan statement.

Researcher check: the two dated amounts and per-seat monthly units support the $10 increase for the same plan. The advert supports only the reported hiring signal. The free-plan correction is consistent with the supplied statement that your plan is paid, but an original approved pricing record is still needed before a verified external correction can be sent. No supplied summary replaces those original records.

Correction request, unsent draft

'To [named AI service or page owner]: Please review the answer to [exact question], observed on [original date], which describes [business name]'s plan as free. That description conflicts with the approved pricing record [attach or link the original record], which identifies the plan as paid. Please correct the free-plan description and cite the current approved pricing record. The relevant answer excerpt is [attach original excerpt], with its cited links [attach links].'

The brackets are missing-input fields, not invented names, dates, links, or prices. Drafting can finish; verification and sending stop until the original answer and approved pricing record are available, the recipient is known, and the person specifically approves sending. The person owns any correction to their own page as a separate action.

No competitor response is recommended from the price and hiring signals alone because the business decision and comparable own offering were not provided. Sources used: the three supplied fictional records. Next owner: the report owner, to retain both price dates, obtain missing search/pricing evidence, complete the correction request, and decide whether to act. Nothing was searched, published, sent, scheduled, or changed externally.

Checklist results

  • pass: Reports the $10 change without mixing annual and monthly prices. — October 1 $29 and October 5 $39 are both per seat monthly for the same plan; result is $10 per seat per month.
  • pass: Keeps the hiring advert separate from a confirmed roadmap. — Enterprise Sales Lead is explicitly a hiring signal, with no confirmed launch, revenue, or financial-health claim.
  • pass: Identifies the wrong price claim and preserves the supplied-observation label. — Paid-as-free description is labeled one supplied AI pricing discrepancy; original answer and metadata remain missing.
  • pass: Makes no private inquiry or automatic website change. — Correction request remains unsent draft, with original evidence and recipient required before specific send approval; no searches or site changes occurred.

Not tested

  • Single-agent staged text handoffs, not separate Bot conversations or enforced security boundaries.
  • No new searches, actual AI-service runs, original-page verification, or correction delivery occurred.
  • Final correction request contains explicit missing fields because original source records and recipient were not supplied.
Task instructions for existing Bots and run limits

Researcher

Keep the existing Bot's role and boundaries. For this crew: Check important claims against their original sources. Check a defined question against the original approved records. Return evidence, source references, and gaps. Label supplied snapshots separately from current live research. Use only the input assigned to you for this task. Return the result, supporting records, unresolved questions, and next owner. A handoff does not approve an external action.

Run limits

Up to five named competitors and five exact questions per approved manual run. No subscription signups or scheduled monitoring.

Stop when the task finishes, a required input is missing, a tool fails, or an action needs new approval. No automatic retries. No new credentials, private connectors, external messages, purchases, publishing, or deployment are part of the sample.

Finished output

A dated comparison with source notes and corrections worth reviewing.

Workflow written by Bot Cabinet. Inspired by The Spy Network, from the supplied @unicodef1wn chart. Read the source notes. See Grok’s official guides for setup and sharing instructions.